tenshi installation fails

[Johan Ehnberg]

Tested on three different Ubuntu 18.04 LTS systems. It seems something is broken in the packaging of tenshi. Installation stalls for several minutes and then fails.

$ sudo apt install tenshi
Reading package lists... Done
Building dependency tree
Reading state information... Done
The following additional packages will be installed:
  libio-bufferedselect-perl
The following NEW packages will be installed:
  libio-bufferedselect-perl tenshi
0 upgraded, 2 newly installed, 0 to remove and 0 not upgraded.
Need to get 36.7 kB of archives.
After this operation, 173 kB of additional disk space will be used.
Do you want to continue? [Y/n]
Get:1 http://helsinki-1.clouds.archive.ubuntu.com/ubuntu bionic/universe amd64 libio-bufferedselect-perl all 1.0-1 [7836 B]
Get:2 http://helsinki-1.clouds.archive.ubuntu.com/ubuntu bionic/universe amd64 tenshi all 0.13-2+deb7u1 [28.8 kB]
Fetched 36.7 kB in 0s (104 kB/s)
Selecting previously unselected package libio-bufferedselect-perl.
(Reading database ... 33410 files and directories currently installed.)
Preparing to unpack .../libio-bufferedselect-perl_1.0-1_all.deb ...
Unpacking libio-bufferedselect-perl (1.0-1) ...
Selecting previously unselected package tenshi.
Preparing to unpack .../tenshi_0.13-2+deb7u1_all.deb ...
Unpacking tenshi (0.13-2+deb7u1) ...
Processing triggers for ureadahead (0.100.0-20) ...
Processing triggers for systemd (237-3ubuntu10.13) ...
Processing triggers for man-db (2.8.3-2ubuntu0.1) ...
Setting up libio-bufferedselect-perl (1.0-1) ...
Setting up tenshi (0.13-2+deb7u1) ...
Adding tenshi system user ...
adduser: Warning: The home directory `/var/lib/tenshi' does not belong to the user you are currently creating.
Adding user tenshi to group adm
Job for tenshi.service failed because a timeout was exceeded.
See "systemctl status tenshi.service" and "journalctl -xe" for details.
invoke-rc.d: initscript tenshi, action "start" failed.
● tenshi.service - LSB: tenshi log monitoring and reporting tool
   Loaded: loaded (/etc/init.d/tenshi; generated)
   Active: failed (Result: timeout) since Mon 2019-03-11 10:06:35 UTC; 206ms ago
     Docs: man:systemd-sysv-generator(8)
  Process: 2792 ExecStart=/etc/init.d/tenshi start (code=killed, signal=TERM)
    Tasks: 3 (limit: 2320)
   CGroup: /system.slice/tenshi.service
           ├─2800 /usr/bin/perl /usr/sbin/tenshi -P /var/run/tenshi/tenshi.pid
           ├─2801 /usr/bin/tail -q --follow=name --retry -n 0 /var/log/syslog /var/log/auth.log
           └─2802 tenshi

Mar 11 10:01:34 rodakorset systemd[1]: Starting LSB: tenshi log monitoring and reporting tool...
Mar 11 10:06:35 rodakorset systemd[1]: tenshi.service: Start operation timed out. Terminating.
Mar 11 10:06:35 rodakorset systemd[1]: tenshi.service: Failed with result 'timeout'.
Mar 11 10:06:35 rodakorset systemd[1]: Failed to start LSB: tenshi log monitoring and reporting tool.
dpkg: error processing package tenshi (--configure):
 installed tenshi package post-installation script subprocess returned error exit status 1
Processing triggers for systemd (237-3ubuntu10.13) ...
Processing triggers for ureadahead (0.100.0-20) ...
Errors were encountered while processing:
 tenshi
E: Sub-process /usr/bin/dpkg returned an error code (1)

[Johan Ehnberg (johan-ehnberg)]

Removing the package after the failed install is not straight forward, here is a workarond:
    sudo systemctl stop tenshi.service; sudo killall tenshi; sudo apt purge tenshi

[George Clark (geoaubuntu)]

Also fails on 16.04. Hanging on multiple installations.

After reboot, tenshi does not start automatically. But manually starting it and waiting past the timeout, the tenshi tasks do start, and tenshi apparently works fine, sending the expected log monitoring messages.

[Launchpad Janitor (janitor)]

Status changed to 'Confirmed' because the bug affects multiple users.

[George Clark (geoaubuntu)]

Here is a startup hang and status on a 16.04 system. Identical behavior on a 18.04 fresh install.

sudo service tenshi start

Job for tenshi.service failed because a timeout was exceeded. See "systemctl status tenshi.service" and "journalctl -xe" for details.

● tenshi.service - LSB: tenshi log monitoring and reporting tool
   Loaded: loaded (/etc/init.d/tenshi; bad; vendor preset: enabled)
   Active: failed (Result: timeout) since Tue 2019-03-12 21:31:32 EDT; 26min ago
     Docs: man:systemd-sysv-generator(8)
  Process: 4484 ExecStart=/etc/init.d/tenshi start (code=killed, signal=TERM)
    Tasks: 3
   Memory: 16.4M
      CPU: 409ms
   CGroup: /system.slice/tenshi.service
           ├─4493 /usr/bin/perl /usr/sbin/tenshi -P /var/run/tenshi/tenshi.pid
           ├─4494 /usr/bin/tail -q --follow=name --retry -n 0 /var/log/syslog /var/log/auth.log
           └─4495 tensh

Mar 12 21:26:32 sys1 systemd[1]: Starting LSB: tenshi log monitoring and reporting tool...
Mar 12 21:31:32 sys1 systemd[1]: tenshi.service: Start operation timed out. Terminating.
Mar 12 21:31:32 sys1 tenshi[4484]: Starting log monitor: tenshi
Mar 12 21:31:32 sys1 systemd[1]: Failed to start LSB: tenshi log monitoring and reporting tool.
Mar 12 21:31:32 sys1 systemd[1]: tenshi.service: Unit entered failed state.
Mar 12 21:31:32 sys1 systemd[1]: tenshi.service: Failed with result 'timeout'.

[George Clark (geoaubuntu)]

Adding --background to the start-stop-daemon command in the init script seems to get around the issue.

start-stop-daemon --background --start -c tenshi --pidfile $PIDFILE --startas $DAEMON -- -P $PIDFILE

This is documented as a "last resort" by start-stop-daemon. Maybe there is a better solution.